rf.combine: Combine Random Forests Ensembles

In rfUtilities: Random Forests Model Selection and Performance Evaluation

Description

Combine two more more random forests models into a single ensemble.

Usage

rf.combine(...)

Arguments

...	two or more randomForest class objects as individual objects or a list containing models

Value

An object of class randomForest

Note

The confusion, err.rate, mse and rsq components (as well as the corresponding components in the test component, if exist) are averaged across ensembles. This is a modification of the randomForest combine function that returns averaged validation statistics

Author(s)

Jeffrey S. Evans <[e-mail]>

See Also

randomForest for randomForest details

combine for original combine function details

Examples

library(randomForest)
data(iris)

c1 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)
c2 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)
c3 <- randomForest(Species ~ ., iris, ntree=50, norm.votes=FALSE)

( class.combine <- rf.combine(c1,c2,c3) )

data(airquality)
set.seed(131)
r1 <- randomForest(Ozone ~ ., data=airquality, mtry=3,
                   importance=TRUE, na.action=na.omit)
r2 <- randomForest(Ozone ~ ., data=airquality, mtry=3,
                   importance=TRUE, na.action=na.omit)
r3 <- randomForest(Ozone ~ ., data=airquality, mtry=3,
                   importance=TRUE, na.action=na.omit)

( regress.combine <- rf.combine(r1,r2,r3) )				   

rfUtilities documentation built on Oct. 3, 2019, 9:04 a.m.